🚇 The Subway Subway

A map of Subway restaurants rendered as a subway system.

# What is a Subway Subway?

As part of my final project for GEOB 372 at the University of British Columbia, I was tasked with creating a map for any topic of my choosing that reflected all that I had learned throughout the course. I opted to do my final project of a "Subway Subway" – in other words, a subway/metro/rapid transit system for the City of Vancouver, with stations based solely on the locations of Subway restaurants.

# Project Background

Ever notice how chain restaurants seem to nail the perfect distribution across a city? That's exactly what kicked off this project. Urbanists across North America had been pointing out something fascinating: if you map out certain restaurant chains, you'll often end up with better transit coverage than actual public transit systems.

While this observation started as a creative thought experiment on social media, it evolved into a serious conversation about transit accessibility in our cities. Vancouver was missing from this analytical party, so I decided to change that.

Here's an article on formerly CityLab, now Bloomberg, detailing the concept in depth.
Link on Internet Archive ↗

Another article on the Curved archive about the concept ↗

# Data Collection Process

# The Game Plan

# The Data Hunt

The foundational elements—base map and transit data—were a breeze thanks to Vancouver's Open Data Portal. Perfect for dropping into ArcGIS or QGIS. But finding every Subway location? That turned into quite the adventure:

  1. Reached out to Subway Canada (radio silence 👋)
  2. Checked Vancouver Coastal Health's Food Asset Map (close, but no sandwich)
  3. Rolled up my sleeves and plotted each location manually in Google My Maps

Pro tip for future mappers: These days, you could tap into OpenStreetMap or business location databases to save yourself some time.

Here's the query in Overpass that would've done this:

node
  [brand=Subway]
  ();
out;

# Data Preparation

Getting from raw data to a workable map was quite the journey. After extracting our initial dataset, I went through several rounds of testing with different software suites and file converters until I found the right ArcGIS-compatible format. (Sometimes the best solutions take a few attempts!)

# Station Analysis

To make the station markers meaningful, we pulled ridership data from TransLink's annual Transit Service Performance Review. This data drove my decision on how each station would be represented visually on the map.

# Design Decisions

When it came to the map's visual approach, I had two paths forward:

I opted for the schematic approach, taking inspiration from TransLink's Fast and Frequent Transit Network map (previously known as the Rapid Transit network map before the 2020 RapidBus Network launch).

# Visual Identity

The design builds on TransLink's established visual foundation, using their standard Meta OT font family. While staying true to their core design principles, we developed our own unique twist with:

For those station markers, I calculated sizes using: